Australian cloud for business software — built and run in Sydney since 2002

NetLeverage
NetLeverage

Scheduled exports from hosted apps to storage the customer controls

Technical article
Hosting

About this article: This article is reviewed against the cited sources. This is source research, not lab validation. Check the stated versions and your own requirements before making changes.

Direct answer: Scheduled exports push or pull business artefacts—database extracts, document bundles, audit logs, configuration snapshots—from a hosted application environment into storage the customer controls (their object bucket, their SFTP endpoint, their secondary site). Done deliberately, exports support reporting, regulator requests, analytics and future provider change without pretending a backup job and a business export are the same product. The design question is what must leave the host’s boundary on a clock, in what format, with what integrity proof—not how to click verify on last night’s backup delivery.

Why it matters

Compliance-minded owners and finance leads often need evidence that sits outside the primary hosting footprint: monthly GL extracts, member registers, transaction logs, or application configuration that proves how reports were produced. If exports exist only as ad hoc support tickets, retrieval during dispute or migration becomes slow and uneven.

Portability benefits accrue when exports are structured and repeatable. A zip of mystery CSVs with no schema document may satisfy a one-off audit but will not help a new provider reconstruct workflows. This article covers business export automation, not the possession check in Verify a backup delivered to S3 or SFTP or whole-estate portability planning when changing provider.

How it works

Export types commonly scheduled from hosted LOB stacks:

Export type Typical content Consumer Integrity expectation
Data mart / reporting Denormalised tables or views BI, accountant Row counts, checksum
Document pack PDFs, images Regulator, legal Manifest of object keys
Audit trail Auth and change logs Security reviewer Timestamped, append-only
Configuration App settings, integration maps DR / migration team Versioned JSON with owners

Delivery channels (generic): push to customer S3-compatible bucket via API or gateway; SFTP put to customer server; synchronise to a customer-managed file share over private network. The Amazon S3 user guide introduction describes S3 as object storage with scalability, security, availability and performance; customers may use another vendor with similar semantics.

Scheduling usually combines an application or database job, a Windows Task Scheduler or Linux cron entry, and outbound network policy. Agree timezone (for Australian reporting, many teams anchor schedules to Australia/Sydney, though your estate may use other zones) and failure visibility: a silent export failure is worse than no export, because leadership assumes possession they do not have.

Integrity and retention (practitioner guidance): pair each drop with a manifest (file name, byte size, SHA-256 hash, record count). Customer-controlled storage should carry a lifecycle policy your legal and finance advisers agree—periods vary by record type and sector. The ATO’s overview of record-keeping rules for business is a starting point for Australian tax records (not a complete retention schedule). Where exports contain personal information, the OAIC summary of Australian Privacy Principles frames collection, use, security and access expectations at a high level—this article is not legal advice; confirm obligations with qualified advisers.

What to choose and why

Export versus backup: backups prioritise consistent restore of systems; exports prioritise human and downstream-system consumption. Some workloads need both, with different schedules and scopes. Do not relabel backup copy jobs as “compliance exports” without defining format and meaning.

Frequency: match the business decision cadence—daily trading extracts, weekly payroll support files, monthly statutory packs. Practitioner guidance: over-frequent full exports can increase egress cost and breach surface (egress pricing is contract- and vendor-specific; model it before scheduling).

Encryption in transit and at rest (practitioner guidance): TLS on HTTPS object APIs and webhooks, with certificate validation against the expected hostname. SFTP rides inside SSH, not inside TLS—require host-key verification (and pinned known hosts where policy allows), not “TLS on SFTP.” Encrypt objects at rest with keys the customer holds where policy demands it. Key custody should be documented in your documented inventory of systems, keys and integrations (or equivalent runbook the customer maintains).

Transport reference: SFTP is an SSH subsystem; see OpenSSH ssh(1) and sftp(1) for client session behaviour, and RFC 4253 for the SSH transport layer.

Least privilege (practitioner guidance): export jobs should use credentials that can write only the agreed prefix in the customer bucket, not administer the hosted production database.

Risks and limits

No export scripts, bucket policies or schedulers were executed in this article (untested). Egress pricing, data residency and privacy obligations depend on contract and data class.

Negotiate export scope with your managed host using plain acceptance criteria.

Sources

Learning notes

  1. Exports answer “what can we hold and read?”—backups answer “can we rebuild the system?”
  2. Manifests and hashes turn opaque zips into evidence.
  3. Customer-controlled storage implies customer key and lifecycle ownership.
  4. Failed exports need alerts, not hope.
  5. Schedule and timezone choices should match Australian reporting cadence.